Enhanced Accuracy in Oil Production Forecasting using Spiking Neural Networks Tuned with Fennec Fox Optimization
Year: 2026
Type: conference-paper
Abstract: For accurate prediction of Oil Production (OP) a new technique called an Efficient Spiking Neural Network (ESNN) optimized Fennec Fox Optimization (FFO) technique is proposed. In the oil industry, where precise projections of future construction levels aid in resource management optimization and operational risk mitigation, the forecasting process is important for efficient management. Using the raw dataset, the input data is fed pre-processing, which contains null values and handle missing values. Transformed and normalised using processed data, it is divided into subsets for testing and training for evaluating the model. These preprocessing procedures guarantee that input data is appropriate for building a reliable forecasting model. Additionally, the feature engineering phase carries out crucial operations like as normalising the data and converting it into a format that improves the prediction. The proposed ESNN, which uses spiking neurones to describe the complicated, time-dependent nature of OP data, is the central component of proposed method. FFO model optimise this network to enhance the methods ability to recognise difficult designs in OP data and create precise predictions about future productions. Using Python software is implemented with Proposed ESNN with FFO MSE of$\mathbf{0. 0 0}$, RMSE of$\mathbf{0. 0 0 5 6}$, MAE of$\mathbf{0. 0 0 1 9}$and high$\mathbf{R}^{\mathbf{2}}$of 97% for OP forecasting.
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W7204773131